Gesture recognition method and wearable apparatus

• US 9,383,824 B2
• Filed: 12/03/2013
• Issued: 07/05/2016
• Est. Priority Date: 09/03/2013
• Status: Active Grant

• Current Assignee: Wistron Corporation
• Original Assignee: Wistron Corporation
• Inventors: Dai, Hsih-Bin, Chen, Chih-Kang, Chen, Wei-Yu, Shen, Chien-Chi, Wang, Kuo-Hsing, Kao, Yi-Sheng
• Primary Examiner(s): Shankar, Vijay
• Application Number:
  US14/094,821
• Publication Number:
  US 20150061994A1
• Time in Patent Office: 945 Days
• Field of Search: 345 7- 9, 345156-162
• US Class Current: 1/1
• CPC Class Codes: G06F 1/163 Wearable computers, e.g. on a belt
• CPC Class Codes: G06F 1/1694 the I/O peripheral being a single or a set of motion sensors for pointer control or gesture input obtained by sensing movements of the portable computer
• CPC Class Codes: G06F 3/017 Gesture based interaction, e.g. based on a set of recognized hand gestures interaction based on gestures traced on a digitiser G06F3/04883
• CPC Class Codes: G06F 3/0488 using a touch-screen or digitiser, e.g. input of commands through traced gestures

Abstract

A wearable apparatus includes a user interface, a motion sensor, a microprocessor and a central processing unit (CPU). In an operation mode, the motion sensor senses a current hand movement trajectory (HMT). The microprocessor generates a velocity curve along a coordinate axis according to the current HMT, and samples the velocity curve according to a first predetermined velocity and a second predetermined velocity to output velocity sampling points. The microprocessor further determines whether a matching number between the velocity sampling points and velocity feature points is greater than a threshold. The current HMT matches a predetermined HMT when the matching number is greater than the threshold. The CPU performs a system operation corresponding to the default HMT when the current HMT matches the predetermined HMT.

Claims

1. A wearable apparatus, comprising:

a user interface;

a motion sensor, configured to, in an operation mode, sense a current HMT;

a microprocessor, configured to generate a plurality of measured velocity curves according to a predetermined hand movement trajectory (HMT), to determine a first predetermined velocity and a second predetermined velocity according to the plurality of measured velocity curves, to generate a plurality of velocity curves along a plurality of coordinate axes according to the current HMT, to sample the plurality of velocity curves according to the first predetermined velocity and the second predetermined velocity to output a plurality of velocity sampling points, and to determine whether a matching number of between the plurality of velocity sampling points matching with a plurality of velocity feature points is greater than a threshold, the current HMT matching a predetermined HMT when the number is greater than the threshold, wherein the plurality of velocity sampling points are intersection points of the velocity curves and the first predetermined velocity, and intersection points of the velocity curves and the second predetermined velocity;

a storage device, configured to store a plurality of velocity feature points of the predetermined HMT, wherein the plurality of velocity feature points are intersection points of the plurality of measured velocity curves and the first predetermined velocity, and intersection points of the plurality of measured velocity curves and the second predetermined velocity; and

a central processing unit (CPU), configured to perform a system operation corresponding to the predetermined HMT when the current HMT matches the predetermined HMT;

wherein the predetermined HMT represents that a user interface changes from having a back side facing a user to having a front side facing the user.

2. The wearable apparatus according to claim 1, wherein the microprocessor generates a first measured velocity curve and a second measured velocity curve according to the predetermined HMT;

sets a first reference velocity and a second reference velocity, the first reference velocity and the first measured velocity curve forming a maximum number of intersection points, the second reference velocity and the second measured velocity curve forming a maximum number of intersection points;

determines a third reference velocity according to the first reference velocity and a precision, and determines a fourth reference velocity according to the second reference velocity and the precision;

calculates a first predetermined velocity according to an average value of the first reference velocity and the third reference velocity, and calculates the second predetermined velocity according to an average value of the second reference velocity and the fourth reference velocity; and

the velocity feature points fall above the first predetermined velocity and the second predetermined velocity.

3. The wearable apparatus according to claim 2, wherein the microprocessor determines a first tolerable range of the first predetermined velocity and a second tolerable range of the second predetermined velocity according to the first reference velocity curve and the second reference velocity curve, and samples the plurality of velocity curves samples within the first tolerable range and the second tolerable range to output the plurality of velocity sampling points.

4. The wearable apparatus according to claim 1, wherein the system operation is activating the user interface.

5. The wearable apparatus according to claim 1, wherein the coordinate axes comprise a first direction axis, a second direction axis and a third direction axis;

the first direction axis, the second direction axis and the third direction axis are perpendicular to one another.

6. A gesture recognition method, comprising:

generating a plurality of measured velocity curves according to a predetermined hand movement trajectory (HMT);

determining a first predetermined velocity and a second predetermined velocity according to the plurality of measured velocity curves;

storing a plurality of velocity feature points of the predetermined HMT, wherein the plurality of velocity feature points are intersection points of the plurality of measured velocity curves and the first predetermined velocity, and intersection points of the plurality of measured velocity curves and the second predetermined velocity;

in an operation mode, sensing a current hand movement trajectory (HMT);

generating a plurality of velocity curves along a plurality of coordinate axes according to the current HMT;

sampling the plurality of velocity curves according to the first predetermined velocity and the second predetermined velocity to output a plurality of velocity sampling points, wherein the plurality of velocity sampling points are intersection points of the velocity curves and the first predetermined velocity, and intersection points of the velocity curves and the second predetermined velocity;

determining whether a number of the plurality of velocity sampling points matching with a plurality of velocity feature points is greater than a threshold, the current HMT matching a predetermined HMT when the number is greater than the threshold; and

performing a system operation corresponding to the predetermined HMT when the current HMT matches the predetermined HMT;

wherein the predetermined HMT represents that a user interface changes from having a back side facing a user to having a front side facing the user.

7. The gesture recognition method according to claim 6, further comprising:

generating a first measured velocity curve and a second measured velocity curve according to the predetermined HMT;

setting a first reference velocity and a second reference velocity, the first reference velocity and the first measured velocity curve forming a maximum number of intersection points, the second reference velocity and the second measured velocity curve forming a maximum number of intersection points;

determining a third reference velocity according to the first reference velocity and a precision, and determining a fourth reference velocity according to the second reference velocity and the precision;

calculating the first predetermined velocity according to an average value of the first reference velocity and the third reference velocity, and calculating the second predetermined velocity according to an average value of the second reference velocity and the fourth reference velocity;

wherein, the velocity feature points fall above the first predetermined velocity and the second predetermined velocity.

8. The gesture recognition method according to claim 7, further comprising:

determining a first tolerable range of the first predetermined velocity and a second tolerable range of the second predetermined velocity according to the first reference velocity curve and the second reference velocity curve;

wherein, the step of sampling the plurality of velocity curves samples within the first tolerable range and the second tolerable range to output the plurality of velocity sampling points.

9. The gesture recognition method according to claim 6, wherein the system operation is activating a user interface.

10. The gesture recognition method according to claim 6, wherein the coordinate axes comprise a first direction axis, a second direction axis and a third direction axis;

the first direction axis, the second direction axis and the third direction axis are perpendicular to one another.